English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Referring Video Object Segmentation (R-VOS) is a challenging task that aimsto segment an object in a video based on a linguistic expression. Most existingR-VOS methods have a critical assumption: the object referred to must appear inthe video. This assumption, which we refer to as semantic consensus, is oftenviolated in real-world scenarios, where the expression may be queried againstfalse videos. In this work, we highlight the need for a robust R-VOS model thatcan handle semantic mismatches. Accordingly, we propose an extended task calledRobust R-VOS, which accepts unpaired video-text inputs. We tackle this problemby jointly modeling the primary R-VOS problem and its dual (textreconstruction). A structural text-to-text cycle constraint is introduced todiscriminate semantic consensus between video-text pairs and impose it inpositive pairs, thereby achieving multi-modal alignment from both positive andnegative pairs. Our structural constraint effectively addresses the challengeposed by linguistic diversity, overcoming the limitations of previous methodsthat relied on the point-wise constraint. A new evaluation dataset,R\textsuperscript{2}-Youtube-VOSis constructed to measure the model robustness.Our model achieves state-of-the-art performance on R-VOS benchmarks,Ref-DAVIS17 and Ref-Youtube-VOS, and also ourR\textsuperscript{2}-Youtube-VOS~dataset.

Towards Robust Referring Video Object Segmentation with Cyclic Relational Consensus